Recently, deep learning has achieved impressive performance on an array of challenging problems, but its success often relies on large amounts ofmanually annotated training data. This limitation hassparked interestin learning from fewer examples. A well-studied instance of this problem is few-shot image classification: learning new classes from only a few representative images.
In addition to being an interesting problem from a scientific perspective due to the apparent gap between the ability of a person to learn from limited information compared to that of a deep learning algorithm, few-shot classification is also a very important problem from a practical perspective. Because large labeled datasets are often unavailable for tasks of interest, solving this problem would enable, for example, quick customization of models to individual user’s needs, democratizing the use of machine learning. Indeed, there has been an explosion of recent work to tackle few-shot classification, but previous benchmarks fail to reliably assess the relative merits of the different proposed models, inhibiting research progress.
In “Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples” (presented atICLR 2020), we propose a large-scale and diverse benchmark for measuring the competence of different image classification models in a realistic and challenging few-shot setting, offering a framework in which one can investigate several important aspects of few-shot classification. It is composed of 10 publicly available datasets of natural images (includingImageNet,CUB-200-2011,Fungi, etc.), handwritten characters and doodles. The code ispublic, and includes anotebookthat demonstrates how Meta-Dataset can be used inTensorFlowandPyTorch. In this blog post, we outline some results from our initial research investigation on Meta-Dataset and highlight important research directions.